THE ECOLOGY OF THE EARED DOVE (ZENAZDA AURZCULATA) IN

R. K. MURTON Monks Wood Experimental Station, N.E.R.C. Huntingdon, England

AND E. H. BUCHER, M. NORES, E. GOMEZ, AND J. REARTES Department of Zoology University of C6rdoba Casilla de Correo 122 Argentina,

In South America, south of Trinidad and other mostly arid and semiarid scrubland, usually islands in the southern Caribbean, the familiar with some trees or patches of woodland, but (Zen&da macroura) of North it avoids the tropical forest areas. Its niche America is replaced by the Eared Dove (Ze- is similar to that occupied by the collared naida auriculata) also known as “Torcaza,” doves of the Old World [the Vinaceous Dove “Mediana,” or “Paloma dorada;” three dis- ( vinacea) , the Mourning Dove tinctive but intergrading races have been de- ( S. decipieny ), and the Pink-headed Dove scribed. The Eared Dove is very similar to (S. roseogrisea) of Africa; the Collard Dove the Mourning Dove in color and pattern but (S. decaocto) of India and since 1930 of differs primarily in having a shorter, less northwest Europe and Britain; and the Turtle graduated tail, heavier bill, more brownish Dove (S. bitorquatu) of Java]. All these doves coloration, and more distinctly yellow-bronze are ground feeders that eat small weed seeds. display plumage on the neck; these two doves In South America Z. auriculata was found in comprise a super-. There are numer- the thornscrub Chaco region of Argentina and ous published references to the Mourning Paraguay, which originally was a mosaic of Dove but very little has been recorded about woodland and grassland maintained as a the biology of its southern congener. Com- stable, climax community by periodic fires parative studies of closely related species are (Odum 1969; Morello 1970). Although very of interest for they often reveal features not common in this favored pulsatile ecosystem, readily seen in isolation. The Eared Dove the species never formed the enormous flocks also poses an economic problem for it has which are still typical of the more xeric ca- prospered with the development of intensive atinga region of Brazil (fig. 1). There the arable farming and thrives on corn and sor- enormous flocks of Eared Doves are nomadic, ghum crops, the latter eaten from the stand- ranging over very large areas in search of the ing crop from November until March. In seeds of wild grass and other plants, in a man- addition, some farmers have almost stopped ner reminiscent of the Flock Pigeon (Phaps growing millet because the take the histrionica) of the Australian desert and the ripe crop. This is true of wheat also, but to African Weaver Finch. Sandwiched between a lesser extent. In its relationship with agri- the coastal and Amazonian rain forests is a culture, the Eared Dove can be compared corridor of savannas known as the Cerrado, with certain Old World pigeons, particularly which extends from the Chaco to the Caatinga, the Wood Pigeon ( palumbus) , while throughout which the Eared Dove is common its capacity to form enormous flocks for breed- (fig. 1). ing and feeding is reminiscent of the extinct In Argentina, the Pampas region was, until (Ectopistes migratorius) of the end of the 19th century, covered with the eastern United States and the Red-billed native grasslands with practically no trees. Dioch (Quelea quelea) an African weaver The Eared Dove was a common, albeit not finch. characteristic, of the region for many The natural habitat of Z. auriculata is migrated in large flocks from the Chaco to feed on thistle seeds which were abundant Sponsored by a fellowship from Conseio National de Investigaciones Cientificas y T&micas de la Rep&lica Ar- on the plains. Hudson (1920) wrote “this gentina and in part from the Ministerio de Agricultura y Ganaderia de C&doba, R. Argentina. is the commonest species of pigeon tribe in 1801 The Condor 76:80-88, 1974 ECOLOGY OF THE EARED DOVE IN ARGENTINA 81

FIGURE 2. Agricultural development in the thorn scrub belt of Argentina has created a patchwork mosaic of feeding and breeding habitat ideally suited to Zenaida auriculata; here the species has become a pest.

south. It is locally referred to by Argentine botanists as “Espinal.” This transition zone agricultural development is expanding and creating a mosaic of croplands and patches of secondary thornscrub, primarily of Pro- sopis and sp. (fig. 2). Here 2. auri- culatu has become a major pest of sorghum, FIGURE 1. Vegetation map of South America. one of the most suitable crops for the area, The Eared Dove favors thorn scrub regions repre- sented by the “Caatinga” of Brazil (1) and the and one which was introduced in the early “Chaco” of Argentina (2) and extends throughout 1950s (Bucher 1970). Also in the arid valleys

the savanna region. the “Cerrado” 3).,, which links y I C of Venezuela and , where irrigation these two regions. Eared Doves are also abundant schemes are enabling sorghum to be planted, in the savanna region of Venezuela and are common 2. uuriculata does cause damage and is be- in the Pampas of Argentina (4). coming a serious problem (De Grazio 1970). the Argentine country. . . . In autumn they RESULTS often congregate in very large flocks, and are FOOD AND FEEDING HABITS sometimes observed migrating, . . . autumn Doves were collected at Piquillin, 60 km ENE and winter movements are very irregular, of C6rdoba (31”S), Argentina, and at Villa and apparently depend altogether on the sup- Ascasubi, 160 km SE of C6rdoba. Monthly ply of food. When the giant thistle has cov- samples of the crop contents were mixed, ered the plains in summer incredible numbers dried, sorted, and weighed to give a gravi- of Torcasas appear later in the season, and metric measure of the diet (table 1). Seeds usually spend the winter on the plains, con- of wild plants were represented in only small gregating every evening in countless myriads quantities, but those found gave some indica- wherever there are trees enough to afford a tion of the natural food preferences of the suitable roosting-place.” Hudson was refer- species. One of the most important was ring to localities not far from Buenos Aires. Echinochlou colonum, a grass which occurs Almost the whole Pampas was ploughed for in savanna regions throughout the world, in agriculture between 1870-1900 and today, areas of damp ground, or following the sea- although the Eared Dove is still common, sonal rains. It is an important source of food the abundant autumn concentrations have of Queleu queka in Africa (Ward 1965). vanished with the loss of thistles. At present, Another grass species, included with Echino- surrounding the Pampas in a great arc stretch- chloa because it is difficult to distinguish, ing from Uruguay and Rio Grande in Brazil, was Seturia pumpeuna. Both of these grasses inland to C6rdoba and back to the coast at produce seeds about l-2 mm in length that Bahia Blanca, Argentina, is a belt of scrub weigh about 1.0 mg. Amarnntlzus spp. (fam- country. This forms an ecotone between the ily Amaranthaceae closely related to Cheno- Pampas and Chaco to the west and between podiaceae) and Chenopodium album (fat the Pampas and Patagonian desert to the hen), introduced from Europe, comprised 82 R. K. MURTON ET AL.

grown for alcohol and starch production or as components for feeding stuffs. The birds collect the seeds from the growing plants and, thereafter in April and May, from the stubbles. As this food supply is depleted, the birds exploit a wider variety of foods, each of which contributes a relatively small share to the total diet. Thus from July to Sep- tember seeds of wild plants feature promi- nently in the daily diet. A second group of sorghums has small elongated seeds, and is grown as a forage crop. S. almum is a peren- nial used entirely to provide pasturage; S. halepense is also a perennial which has be- come wild. It is a pest, and invades many

FIGURE 3. Sequence in which various cultivated fields where it may be grazed, particularly seeds feature in diet of Eared Doves in Argentina. at Piquillin where the emphasis is on cattle- Based on percentage by weight of crop contents of grazing. Wild sorghums (usually a mixture shot birds. See also table 1. Solid dots and lines = of the above two species and their hybrids) cultivated sorghum; open dots and lines = wild ripen at the same season as the cultivated sorghum; solid dots and dashed lines = millet; open dots and dashed lines = wheat; crosses and dotted varieties, usually between February and May. lines = weed seeds. They become important constituents of the doves ’ diet in August and September if they have not been harvested and the seeds still the other major seed foods. Seeds of wild remain on the plants or ground. As the stub- plants taken in smalIer quantities were Croton bles of cultivated sorghums are ploughed, sp., Euphorbia sp. (Euphorbiaceae), and the birds change to this alternate food source. Argemone suhfusiformis (Papaveraceae). Cro- Insects were recorded only in September ton sp. is also a favorite food of Z. auriculata and comprised aphids from a wheat crop that in Brazil (Von Ihering 1935; Aguirre 1964). had been sprayed with insecticides. Perhaps The natural diet of the Eared Dove therefore the aphids resembled seeds. Snails were appears to be mostIy the small seeds of an- eaten during the breeding season, possibly nual plants which can be collected from the for the calcium content of their shells. ground. These feeding habits are typical of other Zenaida and Streptopelia doves, which FOOD AND BREEDING often eat the same or closely related seeds. The climatic conditions in the study areas For instance, in Europe the Turtle Dove (meteorological data from Pilar, 70 km E of (Streptopelia turtur) specializes on grass Cordoba) result in a semi-desert vegetation. seeds and such small seeds as Chenopodium The rainfall is about 600-700 mm per annum (Murton et al. 1964). of which 80% is concentrated during the Today, most of the diet of the Eared Dove October-March period; August and Septem- in central Argentina is comprised of the ber are the driest months and most trees shed seeds of cultivated plants, particularly Sor- their leaves at this time. Temperature ranges ghum, wheat ( Triticum aestivum) , millet are 17-32°C ( mean min/max) in January ( Panicum miliaceum) and, to a lesser degree, and 4-17°C in mid-winter (July). Wheat is maize (Zea mays), peanuts (Arachis hypo- planted in May or June and is half-grown gaea), and sunflower ( Helianthus annuus) by the dry season. The first rains stimulate (table 1). Wheat is the first crop to ripen in rapid growth and sorghum is then planted. November or December and it may constitute The rains come as heavy showers interspersed over 80% of the diet in November. It is fol- with sun and the cultivated crops and wild lowed by millet, which ripens and is eaten in seeds ripen rapidly. Grass seeds and other January, and then sorghum in February or natural vegetation are at peak productivity March (fig. 3). D uring the crop analyses, those in March and April. Scattered comments in sorghums with the big round seeds which are old literature indicate the likelihood that, grown for their grain “sorgos graniferos” were prior to agricultural development in the late grouped to include Sorghum caffrorum, S. 1800s the Eared Doves began breeding some- saccharatum, and S. sudanense in their many time between October and the end of Febru- varieties and hybrids. These sorghums are ary, continuing until April; that is, breeding ECOLOGY OF THE EARED DOVE IN ARGENTINA 83

TABLE 1. Diet of Eared Dove as determined from crop contents. -. % composition of crop contents by weight

Food Jun Jul A11g SeP Ott NW DW Jan Feb Mar Apr May -. ~~ Cultivated sorghnm 71 34 27 17 14 3 6 11 78 64 55 57 Wild sorghum 2 20 37 21 5 0 0 3 6 3 1 1 Maize 11 17 15 13 14 2 1 0 1 12 11 14 Millet 0 0 0 0 0 0 35 80 2 4 2 0 Peanut 9 14 7 10 26 2 0 0 0 7 24 19 Wheat 0 0 0 0 2 82 21 0 0 1 0 0 Other cultivated 12 0 13 3 5 12 4 14

Total cultivated ( % ) 94 87 86 62 64 92 68 95 89 95 94 95

Chenopodium sp. and Amaranthus sp. 2 7 8 14 11 0 6 0 1 0 2 1 Echinochloa colonum and S&aria pampeana 3 6 6 21 17 0 0 1 3 2 3 2 Other wild seeds 1 0 T 2 5 4 24 1 : 0 1 1 Other unidentified 0 0 0 0 3 4 2 3 3 0 1 Snails shells and insects 0 0 0 10 0 0 0 T T T T

Total weight food (g) 1125 521 759 419 459 495 604 544 625 707 747 833 No. birds examined 159 79 106 91 125 90 97 111 112 111 150 162 Amount food per bird (g) 7.1 6.6 7.2 4.7 3.7 5.5 6.2 4.9 5.6 6.4 5.0 5.1

coincided with, or else followed, the summer cereals were less easily found than at Asca- rainy season when seeds became abundant. subi. This regime is still typical in uncultivated The diet of nestlings was similar to that of parts of C6rdoba province where the birds adults although some small differences are live more or less solitarily and never breed apparent in the data in table 2. The nestlings during the winter (Bucher, pers. observ.). received smaller quantities of the larger seeds, In Brazil, the birds breed mostly during the including sorghum, and proportionately more rainy season from April until June (Aguirre small seeds, particularly seeds of wild plants. 1964). In Argentina, the advent of pulse crop The adults recorded in table 2 were collected (edible seed crop of leguminous plants) ara- in the same nesting locality as the chicks, ble farming has extended the breeding season whereas those recorded in table 1 were shot to August or September, depending on the flying to roost or near their breeding stations availability of sorghum. Thus during the and so were a different sample. period April to November, assuming that At hatching, pigeon squabs are fed entirely droughts do not intervene to cause a crop fail- on crop-milk, but as they grow older, the par- ure and bring an end to breeding activity, ents feed them increasing quantities of foods the birds can obtain a succession of cultivated which they collect from the fields (table 3). foods (fig. 3). There is usually a bimodal However, the adults apparently exercise some pattern to the breeding season for activity discrimination in the foods they bring to the virtually ceases in June. This is seen in fig- nestlings, regurgitating only small seeds and ure 4 which shows the number of nests with milk initially and feeding larger seeds as the or young and the total occupied per squabs grow (table 3). Such discrimination hectare and is based on approximately weekly could depend on some manner of selective searches of two nesting localities. regurgitation, the adults “holding back” large The adults cease breeding when they can foods or those not entering the squabs’ gape, still obtain cereal seeds, judged by the crop but this seems unlikely. Instead, it is prob- contents of collected samples (fig. 3), but able that the adults collect only appropriate probably food supplies become restricted at foods. This implies that there is some tem- this season (see below). Some birds lose poral mechanism which ensures that the food their reproductive capacity during the period eaten by the adults for themselves passes to of short “winter” days (see below). The the gizzard, whereas seeds for the nestlings breeding season was longer with more nest- are retained in the crop. For some species ing activity during the winter at Villa Asca- the “milk cells” are sloughed from the crop subi than at Piquillin; cattle raising predomi- wall at specific times of day (Murton 1965) nated in the latter locality so that cultivated and it has been demonstrated that in some 84 R. K. hlURTON ET AL.

TABLE 2. Diet of nestling and adult Eared Doves as determined from crop contents.

% composition of crop contents by weight

January 1970 April 1970 June 1970 September 1970

FO0d Nestlings Adults Nestlings Adults Nestlings Adults Nestlings Adults _~ Gultivated sorghum 0 5 50 57 61 73 45 69 Wild sorghum 9 0 2 1 0 0 4 3 Maize 0 3 2 3 13 8 6 11

PeanutMillet 440 330 212 274 5 130 71 1: Wheat 0 3 0 0 0 0 0 0 Other cultivated 0 6 6 1 1 1 4 1

?%a1 cultivated 53 50 83 93 85 95 67 91 _ Chenopodium sp. and Amaranthus sp. seeds 2 9 7 4 1 0 10 0 Echinochloa colonum and Setaria pampeana 0 0 7 1 9 5 20 8 Other wild seeds 45 40 1 1 0 0 1 0 Other 0 1 2 1 5 0 2 1

Total weight food (g) 101 305 44 373 81 508 182 390 No. birds examined 50 47 62 71 46 86 85 85 pigeons milk production depends on a cir- The Eared Dove normally lays two eggs cadian timing mechanism (Hoffman 1969). and like some other pigeons studied is prob- Conceivably, when the crop becomes filled ably a determinate layer (Poulsen 1953). with milk, the field food collected by the adult Clutches of a single could result from is retained and is then carried to the chicks. partial predation and clutches of three eggs result from a second female depositing eggs BREEDING BIOLOGY in the wrong nests in the exceptionally densely Nests are built in thornscrub at a height cor- crowded nesting colonies. In a sample of 417 responding to the maximum cover although clutches at Piquillin, 77 were of single eggs sometimes nests are built on the ground. In of which 18% hatched; of 622 eggs in clutches lsrazil, birds seem to nest mostly on the of two, 55% of the eggs hatched; and of 87 ground in the shelter of the thorny scrub eggs in clutches of three, 47% hatched. In formed by various bromeliads (pineapple toto, 51% of all eggs laid hatched and 45% group) (Aguirre 1964). Solitary nesting is of the chicks were fledged. If the partial usual throughout large areas of the birds ’ success of whole clutches is considered, then range, where the species occurs singly or in 47% of those laid produced at least one chick small parties, but in the cultivated regions of to fledging. Some slight seasonal variations Argentina enormous colonies of up to l-5 in breeding success were detected; in particu- million individuals can occur. lar, success was low in November and De- cember when few birds were nesting (table 4). Presumably, the doves experienced diffi-

TABLE 3. Percentage by weight of seeds and adult crop milk in the diet of nestling Eared Doves.

% by weight Total Age of weight NO. nestling Seeds SWXIS Seeds birds (days ) Milk < 3 mm 3-4 mm > 8 mm cc%% examined

o-1 100 0 0 0 ? 8 1-2 47 28 25 0 13.3 16 2-3 35 32 33 0 33.8 18 3-4 18 34 48 0 23.9 9 4-5 17 37 38 8 35.8 12 FIGURE 4. Breeding season of Eared Dove in 5-6 5 24 41 30 40.2 14 1970 at two localities in Argentina based on number 6-7 6 36 ;i: 25 24.2 7 nests containing eggs (solid line) or young (dotted 7-8 6 24 18 27.4 6 line). Upper graph data from Piquillin, lower graph 8-9 3 13 75 9 22.3 4 from Villa Ascasubi. ECOLOGY OF THE EARED DOVE IN ARGENTINA 85

700 r 46 b5 64 60 80 78 b4 f54 50 32 20 49 49 44 43 to fully recrudesced throughout the year, so that individuals with spermatozoa could be found in all months. Of a sample of 37 birds collected throughout the year, the testes of one contained secondary spermatocytes, one had spermatids, and the rest contained sper- /t- matozoa. Testes weighing as little as 200 mg contained spermatozoa. Considerable vari- ability was noted between individuals and probably some of the sample detailed in fig- ure 5 included young birds coming into breed- ing condition for the first time. These could not be distinguished by plumage character- istics, but it is known from cage studies that an individual can breed 4-5 months after being fledged (Bucher, pers. observ.). Many individuals begin to breed in the same year as they are born, and similarly, breeding be-

,ool _I _I- L-I havior has been recorded in juveniles of 2. MJJASONDJF MAMJJAS macrowa (Irby and Blankenship 1966). 1970 1971 In any year the minimum mean testis FIGURE 5. Seasonal variation in mean paired tes- weight was noted in the winter months of tes weights (mg 2 S.D.) of Eared Doves in Argen- May-July when daylength is only about 10.25 tina (lat. 32”s ). The number of birds in each monthly sample is given. hr, excluding twilight. A degree of photo- responsiveness was probably involved al- though probably most of the reduction in culty in finding sufficient food during the lat- mean testis size depended on the appearance ter part of the dry season. Those birds which of first-year birds in the breeding population, did manage to breed were doubtless depen- as noted above. Thus most birds continued to dent on such cultivated crops as wheat and exhibit some spermatogenetic development millet (fig. 3 and table 2). Breeding success and presumably could rapidly become fully was also reduced in May at the end of the active. Indeed, the beginning of breeding main nesting season and at a time when the can be very sudden in the Eared Dove, in- gonads of many birds became partly regressed. dicating that the birds are in reproductive Eggs and young may be eaten by various readiness and respond as soon as suitable predators. The opossum ( Didelphis axarae ) , environmental conditions occur. This is a known collectively with related species as characteristic adaptation of many tropical “Comadreja,” is a medium-sized arboreal mar- species, including nomadic ones that must supial which lives in tree holes in scrub areas be ready to take advantage of the appearance and forages on eggs and young in the colonies. of suitable food supplies following rains. A According to local residents, much egg pre- similar pattern of testicular development has dation is caused by a Guira Cuckoo (Guira been found in several Australian pigeons guira) but this has not been confirmed. Many (Braithwaite, pers. corn.). The heaviest tes- scavengers congregate at the breeding colo- tes were noted in a subject killed in March nies, particularly the Red-backed Hawk (989 mg paired weight). Individuals with ( Buteo polyosoma) called Aguilucho comun testes weighing 856, 905, and 802 mg were and the Chimango caracara (Milvago chi- recorded in May and June 1970 and May mango) called Chimango, the latter fre- 1971, respectively, that is, even in those quently was seen attacking nestlings. months when the mean testicular weight was Seasonal changes in gonad size were rela- reduced, there were individuals with fully tively slight compared with those noted in recrudesced organs. However, in July 1971 temperate species. There was only a twofold no bird had testes exceeding 460 mg. difference between the maximum and mini- The ovaries varied more in weight than mum monthly mean testes weights (fig. 5), the testes. The gonad weight of the female whereas reproductively active testes of Wood can exhibit considerable variation depending Pigeons in England exhibit a 42-fold weight on whether she has just become paired, is increase over the regressed condition. The ready to lay, or is caring for eggs or young. testes of most Eared Doves remained partially For this reason seasonal changes in ovarian 86 R. Ii. MURTON ET AL.

TABLE 4. Breeding success of Eared Dove in 4 . Argentina (data for breeding season November 1969- May 1970). 43 . I . . % of clutches % $t,$ss 33 -- 70 y;i;ggs from which No. of which at least one 2 eggs which nestlings nestling fledged / Month laid hatched fledged (No. clutches) / 23 ..A ...... ” ...... , / / November 37 14 0 0 (18) 18' ...... I... December 173 15 12 13( 101) / 13 * ...... ’ . January 104 67 64 69 (55) / / February 162 67 61 65 (83) I . --. . . ..I ...... , . . . March 167 65 63 67 (84) April 74 61 55 54 (41) OCT NoI/ DEC @N FEB MAR AFff MAY JUN JUL WG SEP May 69 52 35 37 (35) Totals 786 51 45 47( 417) FIGURE 6. Primary molt score of Eared Dove near Cbrdoba, Argentina. Large dots = 5 individu- For variations in per cent of eggs from which nestlings fledged als, small dots = single individuals, sample size X”@ = 165.6; P < 0.001. given by numbers. Some lack of synchrony in the stage of molt was noted throughout the population and so the regression is based on the data plotted weight cannot be distinguished; monthly between the dotted lines. The regression line is y = means are given in table 5. -1.0 + 4.2x (month), (the months being numbered from 1 = October to 10 = July); ran = 0.813; P < MOLT 0.001. Many doves were captured which were not The progress of the primary molt is shown molting and these have been omitted. The percent- age of birds not in process of molt each month was in figure 6. Feathers were noted as being Oct.-28% (i.e., 53 birds were molting and are in- old, in process of replacement, or new, but cluded in the figure and 21 were not and are omitted intermediate categories were not recorded. to give 21,/74 X 100 = 28%); Nov.-25%; Dec.- Accordingly, a score of zero was allocated to 36%; Jan.-34%; Feb.-30%; Mar.-46%; Apr.- an old feather, 3 to one being molted, and 5 48%; May-76%; Jun.-92%; Jul.-55%; Aug.- 10%; Sept.-34%. for a new feather, thereby giving a potential score of 50 for a bird which had replaced all 10 primaries. The score can be doubled since at any one time but there was a clear trend the primaries on each wing are molted sym- toward synchrony. In May and June most metrically (two at a time) beginning with birds were molting but many had finished in the innermost and proceeding sequentially July and nearly all by August. Molt appeared to the tenth. The population contained indi- to begin again in October or November, at viduals which were at different molt stages the end of the dry season, for most birds al- though a few individuals were molting the innermost primaries in September while others TABLE 5. Seasonal variations in mean weight * SD did not start until April. Judging from the of ovaries of Eared Dove. pattern of the molt score, it seems likely that

No. birds an individual bird requires about 10 months Year/Month Range examined to replace all primaries, that is, the regression 1970 of molt score on month increases by 4-5 May 121 c 90 20-295 17 points per month (see fig. 6). June 234 e 293 43-1400 33 Body molt occurred coincidently with, and July 615 -c 712 20-2800 53 occupied about the same period as, the pri- August 423 -r- 521 30-2400 62 mary molt, but is was not studied in detail. September 258 -c 428 30-2305 48 - October MOVEMENTS November 334 It 415 43-1840 47 December 100 2 54 20-270 69 The nesting location can be used as a post- 1971 nuptial roost. Subsequently the birds move January 492 & 625 31-2438 51 to other traditional sites as they roam the February 283 & 418 50-1980’ 55 countryside in enormous flocks seeking food. March 361 -c 460 50-1861 59 During the autumn (April-June), a black April 378 & 553 20-1904 38 cloud of 100,000 birds may descend to feed May 144 -c- 191 30-867 25 June - on a group of grain fields. In 2.5 million hec- July 178 c 247 20-1052 31 tares there were between 15 and 20 roosting August 305 & 497 30-2400 47 places each containing a million or so birds. September 251 & 288 35-980 39 The adaptive value of congregating in enor- ECOLOGY OF THE EARED DOVE IN ARGENTINA 87 mous flocks presumably depends on the facili- percentage of clutches from which at least tation of feeding by local enhancement, the one chick was fledged was 51% (cf. 52% in birds being able to exchange information in the present study). Hatching success (per the manner proposed for various other spe- cent of total eggs which hatched) was 59% cies (Zahavi 1971). During the dry season, (51% for Eared Dove) and breeding success food for the doves may be abundant once (per cent of total eggs giving fledged young) located, as on an unploughed stubble, but was 49% (cf. 45% ). For the Eared Dove we such a good source could be isolated within can infer that 88% of the chicks which many square miles of countryside containing hatched were fledged (i.e., nestling success little food. Several instances have been noted = 352 x 100/399 = 88)) while nesting suc- of the birds returning to traditional roosting cess, which was measured directly in the sites even though the trees have been felled. Mourning Dove, averaged 82%. In such cases the birds may roost on the Whereas the Mourning Dove is a valued ground or in low shrubs even though tall sporting bird in North America, the Eared woodland remains elsewhere in the neighbor- Dove is generally rated a pest species in many hood. Similar behavior has been noted in Z. parts of South America. In Colombia the macroura ( Harris 1961) . Eared Dove is reported to cause much eco- During the winter the number of doves nomic damage by eating the seeds and emerg- in the more humid Pampas increases and it ing cotyledons of soybeans (London0 et al. is likely that a proportion of the population 1972). In the province of Cordoba, Argentina, near Cordoba moves to this area to find a the equivalent of US $15,000 is spent offi- new food source. Few doves breed in the cially per annum in attempts to kill the doves Pampas possibly because there is little wood- and other provinces have similar schemes; land or scrub to provide breeding and roost- about one-third of the expenditure is devoted ing places. to strychnine baits which are distributed round the roosts. Although large numbers DISCUSSION of doves are killed, it seems that the problem In its feeding ecology the Eared Dove clearly is not solved. Moreover, it is not likely that bears a close resemblance to the Mourning the slaughtering of doves in this way will Dove in North America. Thus grain sorghum bring remedial benefit if the lessons learned (Sorghum vu&are), Johnson grass (S. hale- from other species, including Quelea, are at pense), one seed croton (Croton monantho- all applicable to the South American scene 1, and annual sunflower (Helianthus (Crook and Ward 1968). spp.) were the major food items of the Mourn- ing Dove in Texas, while such wild seeds as SUMMARY those of Euphorbia spp., Amaranthus spp., The ecology of the Eared Dove was studied Panicum fasciculatum, and other grass seeds near Cordoba, Argentina, in an area of arable were taken in smaller amounts (Dillon 1961). cultivation interspersed with secondary thorn- The same food items were taken by Mourn- scrub. Analysis of crop contents revealed that ing Doves in Oklahoma; the seeds of Grami- the birds’ main diet by weight was composed neae comprised two-thirds of the diet by of cultivated seeds (85%), particularly sor- volume and sunflower seed also was impor- ghum, wheat, and millet. Their feeding hab- tant (Carpenter 1971). So, too, in their breed- its made the birds pests of the farmer. “Nat- ing ecology the two species are very close, ural” foods were grass and wild seeds, with preferring a similar range of nesting habitats Chenopodium, Amaranthus, Echinochloa, and and having virtually identical nests and eggs. Setaria spp. being important constituents of In its southern range, where it is resident, the diet. The food of nestlings was similar. the Mourning Dove has an extended breed- In uncultivated areas the breeding season ing season. Active nests were found in a11 began variably between October and Febru- months of the year except November and ary and extended until April, thereby coin- December in Texas ( ca. 33”N), but the main ciding with and following the summer rains. breeding season was from early March until Arable farming has enabled the breeding sea- early September (Swank 1955). Also in son to begin in August or September so that California (ca. 37”N) nesting lasted from reproduction occurs during about 10 months. mid-March until late September (Cowan Seasonal changes in gonad size and histology 1952). McClure (1950) examined the breed- are described. The molt takes 10 months, ing success of the Mourning Dove in Iowa, generally starting in October and ending in Nebraska, and California and the average July, so that one pair of primaries is changed 88 R. K. MURTDN ET AL.

per month. Rut there is variability in the HOFFMAN, K. 1969. Zum Tagesrhytmus der Bruta- population and some individuals do not be- blosung beim kaptaubchen (Oenu cupensis L. ) und bei anderen Tauben. J. Ornithol. 110:448- gin molting until April. 464. Of the eggs laid, 45% give rise to fledged Hunso~, W. H. 1920. Birds of La Plats. Vol. II. young, while from 52% of the clutches laid Dent and Sons, London. at least one chick is reared. Some causes of InBY, H. D., AND L. H. BLANKENSHIP. 1966. failure are mentioned. Breeding behavior of immature Mourning Doves. The discussion briefly compares the feed- J. Wildl. Mgmt. 30:598-603. ing and breeding ecology of the Mourning LOND~NO, V., D. ELIAS, G. VALENCIA, AND P. W. WORONECKI. 1972. Informe preliminar sobre Dove in the southern United States and men- 1a incidencia de torcaza naguiblanca (Zenuidu tions the differing economic status of these uuriculuta) u su relation con problemas de d&ro two allopatric species that are obviously geo- a algunos cultivos en al valle de1 Cauca, Colom- graphical representatives of a common ances- bia. Inst. Colombiano Agropecuario. 11 p. tral stock. MCCLURE, H. E. 1950. An eleven-year summary of Mourning Dove observations in the west. LITERATURE CITED Trans. 15th N.A. Wildl. Conf. 1950:335-346. MORELLO, J. 1970. Modelo de relaciones entre Acumnn, A. 1964. As avoantes do nordeste. Estud. pastizales y lenosas colonizadoras en el chaco Tecnico No. 24. Ministry of Agric. Brasil. 47 p. argentino. IDIA, Buenos Aires 276:3152. Rio de Janeiro. MUHTON, R. K. 1965. The Wood-pigeon. New BUCHER, E. H. 1970. Consideraciones ecologicas Naturalist Monogr. Collins, London. sobre la paloma Zen&u auricuZuta como plaga MUI~TON, R. K., N. J. WESTWOOD, AND A. J. ISAAC- en Cordoba. Direction Provincial de Asuntos SON. 1964. The feeding habits of the Wood- Agrarios. Serie Ciencia y Tecnica, Cordoba pigeon Columba pulumbus, Stock Dove C. oenas l:l-11. and Turtle Dove Streptopeliu tutiur. Ibis 106: CARPENTER, J. W. 1971. Food habits of the Moum- 174-188. ing Dove in northwest Oklahoma. J. Wildl. Onuxf, E. P. 1969. The strategy of ecosystem de- Mgmt. 35:327-331. COWAN, J. B. 1952. Life history and productivity velopment. Science 164:262-270. of a population of western Mourning Doves in POULSEN, H. 1953. A study of incubation responses California. California Fish and Game 38:505- and some other behaviour patterns in birds. 521. Videns. Medd. Dansk Naturh. Foren. 115:1-131. CROOK, J. H., AND P. WARD. 1968. The Quelea SWANK, W. G. 1955. Nesting and production of problem in Africa, p. 211-229. In R. K. Murton the Mourning Dove in Texas. Ecology 36:495- and E. N. Wright [eds.] The problems of birds 505. as pests. Academic Press, London and New VON IHEHING, R. 1935. La paloma Zen&u auricu- York. l&z en el nordeste de1 Brasil. El Hornero 6:37- DE GMZIO, J. W. 1970. Bird damage problems 47. in Latin America. Proc. 4th Vert. Pest Control WARD, P. 1965. Feeding ecology of the Black- Conf. West Sacramento, California, p. 162-167. faced Dioch Quelea queleu in Nigeria. Ibis 107: DILLON, 0. W. 1961. Mourning Dove foods in 173-214. Texas during September and October. J. Wild]. ZAHAVI, A. 1971. The function of pre-roost gather- Mgmt. 25:334-6. ings and communal roosts. Ibis 113:106-109. HARRIS, S. W. 1961. Migrational homing in Mourning Doves. J. Wildl. Mgmt. 25:61-65. Accepted for publication 15 February 1973.